Table 2 Model performance metrics of ML models in validation data set (95% CI)

LR

0.857 (0.826–0.887)

0.662 (0.597–0.727)

0.817 (0.791–0.842)

0.578 (0.512–0.643)

0.675 (0.603–0.737)

0.858 (0.828–0.883)

0.901 (0.883–0.916)

0.623 (0.561–0.672)

GNB

0.772 (0.734–0.808)

0.480 (0.415–0.552)

0.695 (0.665–0.722)

0.403 (0.353–0.452)

0.751 (0.695–0.806)

0.679 (0.641–0.714)

0.904 (0.884–0.919)

0.525 (0.476–0.574)

CNB

0.834 (0.801–0.867)

0.621 (0.554–0.691)

0.820 (0.797–0.846)

0.602 (0.531–0.666)

0.584 (0.519–0.652)

0.889 (0.865–0.909)

0.881 (0.862–0.897)

0.593 (0.533–0.654)

SVM

0.863 (0.834–0.890)

0.678 (0.624–0.736)

0.800 (0.772–0.826)

0.536 (0.485–0.592)

0.787 (0.729–0.839)

0.804 (0.774–0.831)

0.929 (0.912–0.943)

0.638 (0.589–0.684)

MLP

0.859 (0.826–0.887)

0.687 (0.622–0.744)

0.823 (0.800–0.848)

0.590 (0.519–0.651)

0.680 (0.612–0.745)

0.864 (0.841–0.886)

0.903 (0.885–0.917)

0.632 (0.584–0.686)

AdaBoost

0.851 (0.823–0.880)

0.670 (0.605–0.730)

0.810 (0.785–0.837)

0.563 (0.509–0.620)

0.680 (0.612–0.740)

0.848 (0.822–0.872)

0.902 (0.882–0.917)

0.616 (0.561–0.671)

RF

0.779 (0.745–0.817)

0.547 (0.481–0.622)

0.743 (0.718–0.772)

0.448 (0.392–0.507)

0.635 (0.572–0.701)

0.774 (0.742–0.802)

0.880 (0.858–0.898)

0.525 (0.471–0.573)

Gradient Boosting

0.831 (0.798–0.862)

0.626 (0.564–0.692)

0.802 (0.774–0.829)

0.552 (0.482–0.618)

0.619 (0.555–0.691)

0.855 (0.829–0.878)

0.886 (0.866–0.902)

0.584 (0.529–0.640)

LightGBM

0.815 (0.787–0.846)

0.590 (0.516–0.668)

0.769 (0.741–0.801)

0.489 (0.434–0.542)

0.685 (0.617–0.744)

0.793 (0.763–0.821)

0.897 (0.879–0.913)

0.571 (0.522–0.618)

XGBoost

0.815 (0.779–0.849)

0.601 (0.533–0.668)

0.779 (0.753–0.804)

0.506 (0.445–0.573)

0.650 (0.585–0.724)

0.817 (0.789–0.843)

0.890 (0.871–0.906)

0.569 (0.510–0.620)